5 replies to this topic

[Ritaelyn]

Hi everyone!

 

I'm brand new to astrophotography, planetary imaging, and astronomy in general. I recently purchased a 12" Sky-Walker GOTO Dobsonian telescope, focal length 1,500mm, F/4.91. Accessories I have for it is a Televue 2x Powermate barlow, a ZWO ASI 678MC one shot color camera, and a ZWO ASI Atmospheric Dispersion Corrector.

 

I'd like to share my very first picture of Jupiter with everyone tonight. I shot it at 1024x768, exposure time of 5ms, 186 gain, and it averaged 140 fps. I did a 180 second capture. I used ZWO's included capture tools. My image train was focuser -> 2x barlow. Jupiter was at Zenith so I forgo my ADC for now.

 

The weather sucked badly for this capture. The seeing forecast was 3/5, and oh boy it sure looked like it. There was a constant wave of air infront of jupiter. Jupiter looked like it was under water the entire time.

 

Post processing I used AutoStakkert 4.0 and did 10%, 25%, 33%, and 50% images. I felt 10% turned out the best. I then color corrected it in RegiStax 6 and sharpened it in WaveSharp by playing around with all the sharpness settings until I found something really cool.

 

I was inspired to try to get it to look as close to the images from NASA's Juno Spacecraft - https://www.smithson...raft-180985417/

 

 

Here is my first image ever from my telescope as a complete novice. I'm super excited. My jaw is on the floor with how well my first image turned out. 

 

I'd appreciate kind and constructive feedback. The first immediate issue with the image is there is a verticle line going up and down that I'm not sure how to get out. It's not present on 25%+ stacks, but given the weather conditions the 25%+ stacks don't look nearly as nice as this. I've tried doing some drizziling however whenever I color balance it the sides of jupiter is solid green. My only guess is I should keep at it and definitely get some images when the seeing is 4/5 and 5/5.

 

Thanks!

 

[larrytOMC200]

That is a very good 1st attempt at imaging. As a bit of fun, try using the gaussian filter option on registax, it can give a smoother image. Don't ask me how, I don't know, I sometimes use it on my lunar shots. You will enjoy lunar imaging with your set up, you should give it a try. All the best to you in your new hobby. Larry

[Ritaelyn]

Awesome! I'll try playing around with that filter. Thank you! 😃

[RedLionNJ]

Thank you for a very descriptive recount of how you arrived at that result. And a warm welcome to CN's Major & Minor Planetary Imaging forum.

 

If my first attempt had turned out that good (now many years ago), I'd have been justly thrilled.

 

So what can we offer in the way of advice (which you can either consider or ignore as you wish):

 

Image scale - you appear to be recording at a perfectly reasonable scale for the 678, just a little shy of f/10. But your image on this page looks over-sized - there isn't enough detail present to justify such a large image. Jupiter should be around 300 pixels wide with your setup - it's not clear if you enlarged it at some point in the processing, or if CN did that (as it does, sometimes) when displaying on this page.

 

Those vertical artifacts - I suspect they crept in during alignment/stacking in AS!4. Trust the app to lay out the APs optimally, provided you give it enough in the way of hints. Given the kind of wavy seeing you describe, try to pick an AP size such that AS!4 will pepper the image with between 30 and 40 APs and make sure the "Close to Edge" checkbox is not checked. Multi-Scale is OK.

 

Number of frames to stack - this depends on so many things, including the type of seeing (I think we've all experienced "wavy" too many times) and how perfect the collimation and focus were. Something is a little off, but one can't tell from the result if it's the collimation and/or the focus. Precise collimation (make sure a nearby, unsaturated star shows the Airy disk and concentric diffraction rings, or parts of diffraction rings) makes an incredible difference in high-resolution work. Likewise an exact focus (which can take a bit of time to achieve/verify) makes a significant difference too. If you're focusing with a "hands-on" focuser, this makes precise focus nearly impossible. You really do have to go hands-free. And if the seeing is less than good, verifying collimation is also next to impossible.

 

I think your total capture time is too long for an altaz-mounted scope. Field rotation at this image scale kicks in much sooner than three minutes, I'm afraid. While it can be compensated for (in theory) during processing, it's much, much easier to not have to deal with it in the first place. I'm no expert on altaz planetary captures, but we have plenty of forum contributors who can expand on that aspect.

 

Noise levels - a huge part of planetary imaging is building an appropriate signal-to-noise ratio. You don't have one in that image. The noise is horrendous compared to the signal. Try to include many more frames in your stacks.

 

The color balance in the current version of waveSharp is just as good as that in Registax. In fact, as of this weekend, I would say the newly-released waveSharp v2 makes Registax obsolete. It's still a little sparse on documentation, but that's coming.

 

Wavelets - be more gentle with them. When you think you have things tuned the way you want them, back off a little. When you return later with fresher eyes, you'll appreciate what you did. Human vision has a remarkable propensity for selective blindness and it helps no-one. Until you learn how much sharpening is too much, it's always better to apply too little.

 

I know all these sound like criticisms and they may not even all be valid (e.g. for all I know, you used a great aftermarket focuser and were seeing-limited), but thank you for reading this far. Your initial image shows a lot of potential - you've obviously done research prior to producing it.  And even if you did already read the Planetary AP FAQ (link below), read it again - you'll get a little more out of it after your most recent experience.

 

We look forward to seeing more images from you as you embark upon what may become a lifetime's journey.  Thank you for contributing to CN.

[Ritaelyn]

Thank you for a very descriptive recount of how you arrived at that result. And a warm welcome to CN's Major & Minor Planetary Imaging forum.

 

If my first attempt had turned out that good (now many years ago), I'd have been justly thrilled.

 

So what can we offer in the way of advice (which you can either consider or ignore as you wish):

 

Image scale - you appear to be recording at a perfectly reasonable scale for the 678, just a little shy of f/10. But your image on this page looks over-sized - there isn't enough detail present to justify such a large image. Jupiter should be around 300 pixels wide with your setup - it's not clear if you enlarged it at some point in the processing, or if CN did that (as it does, sometimes) when displaying on this page.

 

Those vertical artifacts - I suspect they crept in during alignment/stacking in AS!4. Trust the app to lay out the APs optimally, provided you give it enough in the way of hints. Given the kind of wavy seeing you describe, try to pick an AP size such that AS!4 will pepper the image with between 30 and 40 APs and make sure the "Close to Edge" checkbox is not checked. Multi-Scale is OK.

 

Number of frames to stack - this depends on so many things, including the type of seeing (I think we've all experienced "wavy" too many times) and how perfect the collimation and focus were. Something is a little off, but one can't tell from the result if it's the collimation and/or the focus. Precise collimation (make sure a nearby, unsaturated star shows the Airy disk and concentric diffraction rings, or parts of diffraction rings) makes an incredible difference in high-resolution work. Likewise an exact focus (which can take a bit of time to achieve/verify) makes a significant difference too. If you're focusing with a "hands-on" focuser, this makes precise focus nearly impossible. You really do have to go hands-free. And if the seeing is less than good, verifying collimation is also next to impossible.

 

I think your total capture time is too long for an altaz-mounted scope. Field rotation at this image scale kicks in much sooner than three minutes, I'm afraid. While it can be compensated for (in theory) during processing, it's much, much easier to not have to deal with it in the first place. I'm no expert on altaz planetary captures, but we have plenty of forum contributors who can expand on that aspect.

 

Noise levels - a huge part of planetary imaging is building an appropriate signal-to-noise ratio. You don't have one in that image. The noise is horrendous compared to the signal. Try to include many more frames in your stacks.

 

The color balance in the current version of waveSharp is just as good as that in Registax. In fact, as of this weekend, I would say the newly-released waveSharp v2 makes Registax obsolete. It's still a little sparse on documentation, but that's coming.

 

Wavelets - be more gentle with them. When you think you have things tuned the way you want them, back off a little. When you return later with fresher eyes, you'll appreciate what you did. Human vision has a remarkable propensity for selective blindness and it helps no-one. Until you learn how much sharpening is too much, it's always better to apply too little.

 

I know all these sound like criticisms and they may not even all be valid (e.g. for all I know, you used a great aftermarket focuser and were seeing-limited), but thank you for reading this far. Your initial image shows a lot of potential - you've obviously done research prior to producing it.  And even if you did already read the Planetary AP FAQ (link below), read it again - you'll get a little more out of it after your most recent experience.

 

We look forward to seeing more images from you as you embark upon what may become a lifetime's journey.  Thank you for contributing to CN.

 

This is very useful feedback! Thank you!

Yes, I did read the planetary FAQ.

 

I had to crop the image from its 16:9 ratio to get the filesize low enough to upload it directly to the forums, which it then zoomed it in more and it's not being displayed at 1:1 scale. It is 347 pixels wide. Is that an acceptable amount of pixels for my 305mm aperture, 1,500mm focal length telescope? Or should I be down sampling it to 305 pixels?

Yes - I used 40 APs on my stacking. As I wrote in the post the 20% make the vert lines go away, but the entire image quality was much worse than 10%. Maybe I could restack it at 15%.

Honestly I think the weather was a huge deal breaker. It was 2-4" FWHM. I couldn't even use my 150x mag (10mm) eye piece to see jupiter that night. So I think it'd go a lot better if I can get 1-2" weather or better.

 

I captured roughly 25,000 frames and took the top 10%, I felt the 20% one was worse. Maybe 20% - 50% will go a lot better in 1-2" or < 1" weather.

 

I also think some of the noise too was over-use of my sharpening. I'll tone it down a lot. Subjectively I like the sharper images but I do think it might be a bit too sharp.

 

I didn't know waveSharp had color balance tools - as a beginner I've been following the FAQ and various tutorials that are using the older tools. Thank you!

 

Speaking of the focuser - I think I need to install the ZWO EAF which will help a ton. I also installed a 1:10 fine focuser knob too, otherwise its the stock crayford 2" focuser. I did the gamma trick per the FAQ to try to get jupiter in focus as best as I could, although I bet more experience will help there!

 

Speaking of other mods and collimation - I made sure the telescope is perfectly collimated to the best of my (beginner) abilities. I bought a howie glatter tublug. I got the telescope so the laser produces cocentric rings both secondary -> primary, and the primary reflects concentric rings. Furthermore doing the star test - defocusing on a bright guide star like vega, etc, shows perfect concentric circles. I also collimate until I have perfect square diffraction spikes on the guide stars too.

 

I even rechecked collimation once I was aligned to jupiter too and it held. A couple weeks ago on the telescope it was losing collimation every time I changed its angle. I fixed this several ways - I tightened the secondary mirror spider - it was loose. The secondary mirror was too low in the telescope so I raised it. The secondary did not look circular - so I rotated it until it looked circular. 

 

I also installed bob's knobs too which helped a ton with the collimation. I also installed Bob's springs for the mirror - that fixed the collimation changing depending on the angle the telescope was looking at. I've also figured out how to secure the mirror screws too which helps a ton, and secure it without having the collimation change as well. 

 

I'd definitely would like to hear more feedback on field rotation. I thought the stacking software adjusted for it and so on. Does anyone else have any inputs on field rotation? Given I'm shooting 140 fps - field rotation most definitely will be sub pixel....

 

I could try to up the framerate more and lower the region of interest. Unfortunately the AZ mount kinda sucks on the skywatcher. Jupiter was bouncing all over the place with a 1024 pix region of interest, which lowers the framerate. I'm not sure if I can get it to track better than that. I'm wondering if I'd be better served with a much sturdier and more precise equitorial mount. 

 

So yeah, I really do think the weather unfortunately played a huge part. I also think I over sharpened the image. Thank you for your feedback.

Edited by Ritaelyn, 08 December 2024 - 01:12 AM.

[Tulloch]

I'd definitely would like to hear more feedback on field rotation. I thought the stacking software adjusted for it and so on. Does anyone else have any inputs on field rotation? Given I'm shooting 140 fps - field rotation most definitely will be sub pixel....

Field rotation is generally handled well by the software, but there's only so much it can do. On my C9.25" SCT I've found AutoStakkert can handle up to about 3 degrees of field rotation before the images start to suffer (see link here). However, larger OTAs are probably more susceptible as the images are larger (assuming the same focal ratio/pixel size value). Field rotation has nothing to do with frame rate, it's about the amount of rotation that occurs between the first frame and the last frame in your video.

 

I could try to up the framerate more and lower the region of interest. Unfortunately the AZ mount kinda sucks on the skywatcher. Jupiter was bouncing all over the place with a 1024 pix region of interest, which lowers the framerate. I'm not sure if I can get it to track better than that. I'm wondering if I'd be better served with a much sturdier and more precise equitorial mount. 

If you are using FireCapture to take the video (which is highly recommended over ZWO's capture software), you should use the "Cutout" box during capture. This will centre the planet in the middle of the captured video and make stacking easier. It wont stop the planet bouncing all over the screen, but at least the video will be stable. 

 

Try to match the framerate with the shutter speed. Remember that you cannot save frames faster than you capture them, if the maximum framerate you can capture is 140 fps, then you should be using an exposure time of (1000/fps = 1000/140) around 7 ms. Capturing at a faster shutter speed is "wasting light", and the camera is capturing frames faster than you can save them. This means you need to use a higher gain which produces more noise.

 

You have done a great job with your first capture session on the planets. With a bit more tweaking here and there and getting some better seeing, you should be able to get some great images of the planets.

 

Andrew

Edited by Tulloch, 08 December 2024 - 03:58 PM.